| Full name | Structurally-compressed Content-preserving Iterative Optimized Network |
| Description | Our research focuses on bridging the performance-efficiency gap in stereo matching by developing SCION-MonSter, a lightweight stereo depth estimation model derived from the state-of-the-art MonSter architecture. |
| Programming language(s) | Python+CUDA |
| Hardware | RTX4090D |
| Website | https://github.com/rayring539/Monsterl-ite |
| Source code or download URL | https://github.com/rayring539/Monsterl-ite.git |
| Submission creation date | 31 Mar, 2026 |
| Last edited | 31 Mar, 2026 |
